Predicting sex from retinal fundus photographs using automated deep learning

Abstract Deep learning may transform health care, but model development has largely been dependent on availability of advanced technical expertise. Herein we present the development of a deep learning model by clinicians without coding, which predicts reported sex from retinal fundus photographs. A...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Edward Korot, Nikolas Pontikos, Xiaoxuan Liu, Siegfried K. Wagner, Livia Faes, Josef Huemer, Konstantinos Balaskas, Alastair K. Denniston, Anthony Khawaja, Pearse A. Keane
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2021
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/c79d88f9f4344e5db04de21644fcea1e
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:c79d88f9f4344e5db04de21644fcea1e
record_format	dspace
spelling	oai:doaj.org-article:c79d88f9f4344e5db04de21644fcea1e2021-12-02T15:43:08ZPredicting sex from retinal fundus photographs using automated deep learning10.1038/s41598-021-89743-x2045-2322https://doaj.org/article/c79d88f9f4344e5db04de21644fcea1e2021-05-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-89743-xhttps://doaj.org/toc/2045-2322Abstract Deep learning may transform health care, but model development has largely been dependent on availability of advanced technical expertise. Herein we present the development of a deep learning model by clinicians without coding, which predicts reported sex from retinal fundus photographs. A model was trained on 84,743 retinal fundus photos from the UK Biobank dataset. External validation was performed on 252 fundus photos from a tertiary ophthalmic referral center. For internal validation, the area under the receiver operating characteristic curve (AUROC) of the code free deep learning (CFDL) model was 0.93. Sensitivity, specificity, positive predictive value (PPV) and accuracy (ACC) were 88.8%, 83.6%, 87.3% and 86.5%, and for external validation were 83.9%, 72.2%, 78.2% and 78.6% respectively. Clinicians are currently unaware of distinct retinal feature variations between males and females, highlighting the importance of model explainability for this task. The model performed significantly worse when foveal pathology was present in the external validation dataset, ACC: 69.4%, compared to 85.4% in healthy eyes, suggesting the fovea is a salient region for model performance OR (95% CI): 0.36 (0.19, 0.70) p = 0.0022. Automated machine learning (AutoML) may enable clinician-driven automated discovery of novel insights and disease biomarkers.Edward KorotNikolas PontikosXiaoxuan LiuSiegfried K. WagnerLivia FaesJosef HuemerKonstantinos BalaskasAlastair K. DennistonAnthony KhawajaPearse A. KeaneNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-8 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Edward Korot Nikolas Pontikos Xiaoxuan Liu Siegfried K. Wagner Livia Faes Josef Huemer Konstantinos Balaskas Alastair K. Denniston Anthony Khawaja Pearse A. Keane Predicting sex from retinal fundus photographs using automated deep learning
description	Abstract Deep learning may transform health care, but model development has largely been dependent on availability of advanced technical expertise. Herein we present the development of a deep learning model by clinicians without coding, which predicts reported sex from retinal fundus photographs. A model was trained on 84,743 retinal fundus photos from the UK Biobank dataset. External validation was performed on 252 fundus photos from a tertiary ophthalmic referral center. For internal validation, the area under the receiver operating characteristic curve (AUROC) of the code free deep learning (CFDL) model was 0.93. Sensitivity, specificity, positive predictive value (PPV) and accuracy (ACC) were 88.8%, 83.6%, 87.3% and 86.5%, and for external validation were 83.9%, 72.2%, 78.2% and 78.6% respectively. Clinicians are currently unaware of distinct retinal feature variations between males and females, highlighting the importance of model explainability for this task. The model performed significantly worse when foveal pathology was present in the external validation dataset, ACC: 69.4%, compared to 85.4% in healthy eyes, suggesting the fovea is a salient region for model performance OR (95% CI): 0.36 (0.19, 0.70) p = 0.0022. Automated machine learning (AutoML) may enable clinician-driven automated discovery of novel insights and disease biomarkers.
format	article
author	Edward Korot Nikolas Pontikos Xiaoxuan Liu Siegfried K. Wagner Livia Faes Josef Huemer Konstantinos Balaskas Alastair K. Denniston Anthony Khawaja Pearse A. Keane
author_facet	Edward Korot Nikolas Pontikos Xiaoxuan Liu Siegfried K. Wagner Livia Faes Josef Huemer Konstantinos Balaskas Alastair K. Denniston Anthony Khawaja Pearse A. Keane
author_sort	Edward Korot
title	Predicting sex from retinal fundus photographs using automated deep learning
title_short	Predicting sex from retinal fundus photographs using automated deep learning
title_full	Predicting sex from retinal fundus photographs using automated deep learning
title_fullStr	Predicting sex from retinal fundus photographs using automated deep learning
title_full_unstemmed	Predicting sex from retinal fundus photographs using automated deep learning
title_sort	predicting sex from retinal fundus photographs using automated deep learning
publisher	Nature Portfolio
publishDate	2021
url	https://doaj.org/article/c79d88f9f4344e5db04de21644fcea1e
work_keys_str_mv	AT edwardkorot predictingsexfromretinalfundusphotographsusingautomateddeeplearning AT nikolaspontikos predictingsexfromretinalfundusphotographsusingautomateddeeplearning AT xiaoxuanliu predictingsexfromretinalfundusphotographsusingautomateddeeplearning AT siegfriedkwagner predictingsexfromretinalfundusphotographsusingautomateddeeplearning AT liviafaes predictingsexfromretinalfundusphotographsusingautomateddeeplearning AT josefhuemer predictingsexfromretinalfundusphotographsusingautomateddeeplearning AT konstantinosbalaskas predictingsexfromretinalfundusphotographsusingautomateddeeplearning AT alastairkdenniston predictingsexfromretinalfundusphotographsusingautomateddeeplearning AT anthonykhawaja predictingsexfromretinalfundusphotographsusingautomateddeeplearning AT pearseakeane predictingsexfromretinalfundusphotographsusingautomateddeeplearning
_version_	1718385836348669952

Predicting sex from retinal fundus photographs using automated deep learning

Ejemplares similares